World of Warcraft’s difficult past, brilliant present and hopefully better future with game designer Matt Goss

WoW past present future

World of Warcraft has had a hell of a year. Pre-Legion it was a bit of a wasteland, bereft of players, stuff to do or updates for months. Legion, and the seemingly-endless flow of stuff that has followed, is amazing – not only in its pure quality, but in the turn around from a team that had lost the trust and faith of its playerbase over the last 18 months. At BlizzCon 2016, they kept up that pace, announcing one major patch, teasing another, and explaining how they was always going to be something new to play for the heavily invested.

For more from BlizzCon 2016, we covered every panel and announcement.

Matt Goss has been through all that. He’s a lead designer on World of Warcraft, working on everything – seriously, he lists professions, overseeing boosts, missions, classes, PvP and more as what he’s tackled just recently. We asked him about that transition from Warlords of Draenor, the launch of Legion, and where it’s all heading, all of which you can read below.

Getting past Warlords of Draenor

World of Warcraft Matt Goss

PCGN: So you guys had quite the challenge ahead of you, right? I’m not going to say Warlords was a mistake or something, but it was certainly poorly received, particularly as it went through its lifespan, but you kinda did it. Like, you nailed it, I would say. I think Legion’s probably the most popular we’ve seen the game be in some time. What do you think led to that? How did you guys go from disappointment/not having a lot of content coming out to this constant process where it’s happening all the time?

MG: I think we look at Warlords as kind of a launching point more than a mistake. We did a lot of things that were really good in Warlords, and we had some missteps. And one of the ones we wish we had done at the time was just being better at getting content out, and having a lot of content for players to play through. So one of the big focuses in Legion was let’s bring it back to the world, let’s make the world cool again, let’s make the world a place that people hang out in, and they progress through storylines like Suramar, they play world quests, they get to see the world as we built it, because it is one of the biggest characters in our game.

We knew that that was a big point on Legion, and then also ‘okay, we want to actually make sure there’s always something for players to play, we need to have content out there, so we can tell our storyline, so we can get those story beats that we really wanna tell, but also so players have something that they can do.’ You saw that right away with 7.1, 7.1 compared to 6.1, it was around the same timing, but we have a dungeon to do, we have a three-boss raid with Trials of Valour to do, that really pushed us into ‘let’s actually make sure this cadence works out’. 7.2, 7.15, there’s a lot of content in those and hopefully we can keep that train going.

Has it been challenging? Was there a point where you were near the end of Warlords and you realised ‘Legion’s not going to come out for another six months, we haven’t got anything in Warlords’ were you just like ‘right, as bad as this is gonna get, we need to create a backlog where we can release 7.1 super quickly after launch and we can immediately have 7.15 on the PTR’?

It wasn’t a trade-off at that point. With Warlords we actually had a few things planned out and Legion was already far along at that point and we said ‘well we really think it’s important to make Legion as amazing as possible’ and it wasn’t until the end of Legion [development] that we were getting… because at some point we lock out right? And we have a little bit of time where we can’t touch the game anymore but we need to have something else to do and at that point was when we said ‘well, we already know what we want to do, we want to do Karazhan and we want to do Trials of Valour, let’s get those in the pipe.’ I think going back we would have liked to have another patch from Warlords, but I think that would have taken away from what Legion is at the same time.

Do you think you’re going to be able to keep up the train? Like The Nighthold isn’t even out yet, Trials of Valour is out in three days, I managed to do Karazhan on the first night it was out but then I was moving house, so I’m like ‘I’ve gotta do how much?’ Do you think you’ll be able to keep up that pace, and do you think that pace is good?

I do think we’re going to be able to keep up the pace. What I think we’ll do is, what we always do is listen to feedback, and we know there’s two sides of the camp, and one is ‘I play WoW all the time, I’m out of things to do, I’ve already done Karazhan twice’ and the other side is like ‘oh my god, everytime I log on it’s like I can’t even finish my Suramar quests because I immediately go into a keystone dungeon, or I have raid night or I have Karazhan to do.’ I think we’re trying to see that right cadence where we can hit both, where a player who is running out of things to do has the next thing to look forward to, and at the same time players who are feeling a little overwhelmed by the content have time to explore it all before we move into the next thing.

How Legion was launched

Legion launch

This is something I want to talk to the Overwatch team about as well, but what was it like on launch night when the servers just went up and they worked, and the reviews start to come in, and the player feedback starts to come in, what was that like after two years of Warlords where you guys took a bit of a kicking, let’s be honest?

We took a bit of a kicking, but the team knew what we were doing. I’ve said this before, I think we didn’t know how good Legion was. I think that sounds like a cop-out but we were so heads down focused on making it better that we weren’t really sure what the reception was going to be. We had a couple of hints early on, we were looking at beta user concurrency. Normally, we actually have to add more people in order to keep concurrency up, so we can get people to test the game. People weren’t leaving, they were playing.

And we were like ‘oh god, okay, this could be really good, let’s keep going and keep going through it’. And on launch day it’s half excitement of ‘oh my god now players are going to be able to do it’ and it’s half terrifying of ‘okay, what have we not done, what did we forget about, is there anything that’s going to cause a problem’ but man, our engineering team knocked it out of the park this time, oh my god. We had so many people playing, and everything was just smooth. Just everything was planned out, best launch we’ve ever had from them, it was phenomenal. I don’t know what kind of magic they used to do that, but it was phenomenal.

I suppose it’s more of an engineering question but do you know why things managed to go so well with the server sharding and all the rest of it?

We had a plan, we had a lot of pre-launch events this time too, and those pre-launch events actually stressed some of that technology, so when we had a lot of people running into Westfall, that was actually using that sharding tech in order for everybody to get their own instance and be able to play through that which led us to fix a lot of problems there before we actually got to live.

And I think we also just worked better with them this time, we were more upfront about like ‘hey, this is how it’s going to play out’ and sometimes they get taken by surprise, which is not our intent, it’s just we’re doing our thing and they’re doing their thing, so we sat down and said ‘this is the way it’s going to play out’ and we found ways to solve the problems and it was just great.

Improving Legion classes, legendaries and more

Legion classes

Ion [Hazzikostas, game director] talked about it, I think it was in one of his Q&As, he said that they regretted how professions worked, but do you know what else there is that you guys feel you could have done better that maybe you didn’t do as well as you could have, and doesn’t – perhaps more importantly- get up to the quality of the rest of Legion?

I think in some ways we’re a lot of times our own worst critic so we have lists of things where we’re like ‘we could have done this better, we probably could have hit this better.’ I think for us it’s trying to look at all the ones we can do something about, and the sooner we can do something about it the better off, so you see it with 7.15 and class updates, that’s our first opportunity after launch.

7.1 is in development before Legion comes out and 7.15 is kind of the first time we have to ‘okay, let’s actually look at classes as a whole, what can we do?’ Maybe it’s not just a number sweep, maybe it’s a ‘let’s take this talent out and change it with a different one’, that’s one example of the thing we’re looking at.

With professions, we’re monitoring feedback on that, I think we want to keep the crafting professions like blacksmithing, tailoring, leatherworking, we want to keep those relevant going forward, make sure that Obliterum doesn’t [get] what happened last time where before raids came out, [crafted gear] was very good, and once raids came out it diminished. We want to try and see if we can solve that problem and come up with a system that lets us do that. It’s not one big thing we look at, it’s just little micro things that we can kind of amp everything up with.

Something that Ion spoke specifically about at the panel was secondary stats as well, and there are times that – I have a feral druid friend and he’s one of the best players I know, and I’ll see stuff drop from him and I’ll be like ‘oh man well done, you got like an 890 Titanforged bit of gear’ and he’ll be like ‘yeah it’s got Haste on it so my 850 crafted shoulders are better’, what are you going to do about that stuff, because obviously that’s just not how it should be, right?

It’s not how it should be. And there’s a lot of problems where, after launch, we realised that we had been solving other problems. This is kind of the quintessential game design problem is that sometimes you focus on one thing and that one thing we tried to solve was as secondary stats ramp up eventually they become so good that they break our classes, and we fixed that… [laughs]

WoW Illidan

That problem rippled through everything else and by the time we really realised it, we didn’t have a fix that was like, this is the golden fix, this is the one that will change everything. This is a huge problem for us, we want to make sure players that are doing new content, especially the first time you step in a raid, you kill the Heroic or Mythic boss for the first time, you want to make sure that that item is good for you, and when you have any doubt whether it is good for you or not, that takes away from the experience, and if it’s really not good for you because you ran it through a bunch of simulations or looked on a bunch of websites with spreadsheets and that determined that it wasn’t good for you, that’s the second moment of ‘okay, why am I doing this?’, that’s a big problem.

So I think 7.15 is a time for us to try to fix that problem. I’m not sure exactly how we’re going to fix it, because it’s a touchy thing to do and we’re trying not to make it have ripple effects on the rest of our game, but it’s something we’re going to fix.

I suppose it is a super challenge as well, because, for example, I play a Prot Warrior, crit is our worst stat I think. You could make crit great for us, and then all my gear is bad, and you don’t’ want to have that problem again. You don’t want to have people log on and be like ‘oh, I’m 20% worse than I was yesterday, that’s cool Blizzard, cheers.’

It’s a lot of things, right? If we do too much we could cause a huge imbalance in the game again too. We’re still tuning things, but where we were at launch and where we are now, we’re in a much better place now, we’d have to re-evaluate all that too. All the choices players have made over the time, they wouldn’t have any trust in items, like ‘my bags are full of rings and necks because I’m terrified of what secondary stat is going to be best or what artifact powers are going to change’ and it’s not a great place to be, and we recognize that, and we’re trying to find the right solve for it right now.

I assume that’s something you do want to have in the game, you want to have something for the player who goes ‘okay that’s 850 and that’s 880 but this is actually better for me’ and so is trinkets where you’re targeting that sort of thing?

We like trinkets because they have effects that do feel different and you feel like there’s more of a choice there, it’s not just the math has said that this one is better. Sometimes we obviously have the ones that the maths are better, like we have trinkets where if you line up with your cooldowns that actually just magically makes them better.

WoW legion itemisation

Then stuff like the Arcanocrystal, right? Where it just had a lot of stat budget.

And then when we realise ‘oh hey, secondary stats are actually better than primaries in some cases, that’s going to be a really good trinket’, those are things that we look at and try to fix. We like having variety, the flamewreath trinket that’s in Karazhan right now is kind of a different playstyle, it lets you toy around with it a little bit. We’d like if that was a choice and it wasn’t just, ‘no, that one’s bad because you have to do something extra to it’, we wanna make sure it feels like ‘okay, if I want to do that gameplay I can take that trinket, if I just want to be passively better, well the Arcanocrystal is the way to go’ when they dominate so much, there’s no play there, it’s just the right answer.

On a similar topic, how do you guys feel about legendaries right now? I have one, where’s my second one, Matt, come on?

You know what? I was really hoping- I didn’t have my legendary at all until Wednesday when I raided, so I was like ‘I can just say I don’t have one either’, and then it dropped and I was like that sense of relief and also ‘damnit’.

So part of the excitement is the rarity, and part of the excitement is not knowing exactly where they’re going to come from, because if you knew exactly what was going to happen it could lead to a more degenerate playstyle. We’d like it to feel like this, ‘okay, I’m playing the game, and then I find one’. That being said, we made some adjustments already where we continued to look at it and made sure everybody has a fair experience and they’re not playing for large periods of time without seeing one but it’s something we’re still looking at. Maybe we’ll have new ways to get them in the future, too. Maybe it’s not just random drop.

Because it’s going to hit the point at some stage where they’re going to be semi-required to be part of the highest tier, and someone who’s coming in and wants to be really hardcore and happens to have not played WoW before and getting involved with Legion and they don’t have 300 days of Emissary quests at that point, so you want to maybe get to a point where you can – maybe not give that person a legendary, but make it very likely for them to be able to get one if they want to pursue one.

Just like we do with catch up mechanics, and we do catch up mechanics for levelling as well, just getting people caught up to current expansions, it’s something we’ll look at for legendaries too. We know if you come in during the 7.2 timeframe and you don’t have a legendary and you’re using Legion launch tuning values for getting one, you’d have a lot of time. We recognize that and we’ll have something that fixes that.

Legion’s secrets and the battle against datamining

Image credit: Wowhead user GeneWow

Were you involved at all with the number of secrets that were in the game when it launched. There’s a couple that really stick out, were you involved with those at all?


Where did all the ideas – particularly for Kosumoth, that was ridiculous. It’s ridiculous that you guys made it, and it’s ridiculous that we found it in like four days.

The best part about that story is that the people that found it weren’t looking for it. They were looking for something else, the Hippogryph. It’s not the hidden that we love, we love the discovery, we love the sense that players don’t know everything about the game, and they haven’t found everything in the game so I think the philosophy is strong. I think as a team just making sure that the actual ways we hide things and the ways that people discover them still feel fun and don’t feel like they just have to do a lot of quest tracking on their characters or read on webpages on how to get something and it feels more like ‘oh, I discovered something myself and because I discovered it I get this reward.’

With that kind of thing and battling against how prolific stuff like Wowhead is. Is it difficult sometimes being like ‘this is a great community tool, we’re so glad that this exists’ but trying to hide stuff from datamining?

It’s difficult, and it’s a struggle too because some of the details that come out end up helping us too – it’s not just the hidden stuff and trying to keep all of our secrets back, but sometimes things come out on webpages that we can actually get feedback on before they come out for real. It’s hard sometimes for players to evaluate something in the game versus just looking at a webpage and reacting to data changes, but it does start conversations earlier so we can kind of get a sense of what we do before the game comes out.

With these hidden ones it is a struggle. It’s something we’re trying to figure out what the right balance is and I think the way to approach it is if the game has a way that I can figure it out, players are more likely to do that. If it feels like I have to be able to look at something because I have no way to know, it can be a challenge, and that’s when they go to datamining websites to figure out what the thing is.

Wowhead interview

I think it is what it is too, players now are more sophisticated, they know more about secondary stat balance. If you think about where this was 5, 6, 7 years ago people kinda had an idea of what their stats were but now it’s like ‘I take my character, simulate the crap out of it, and I get these numbers and I use these numbers and these are my numbers, not even your numbers’. It’s so cool to see that passion but at the same time we have to figure out how to react to it and work to make things better.

How did you build that kind of stuff without it getting all immediately datamined? Was part of it just not implementing it until it was necessary or is there stuff that you can put in with a certain blocker in front of triggers and that sort of thing?

It really depends on the thing, we have some things that we use. I don’t want to spoil too much about how we do it because-

Because then we’ll get past it [laughs].

I mean that’s the thing, it’s like data security, right? It’s a little bit of an arms race. We want to make sure that whatever’s there, it feels like it’s part of the game and we’re not just making it hidden to make it hidden.

How do you keep coming up with that stuff? How do you come up with new stuff we’re not going to see, that you’re trying to hide from us but want us to discover?

We’ll figure it out. We have a couple of new plans. Again, I use the word ‘discovery’ because it’s not really the hidden part that’s cool, it’s the fact that players can discover it on their own. I think those kinds of systems are really powerful and they drive these huge communities. Watching the Kosumoth forum threads, and the people just trying to figure it out and trying to find all these bits and pieces, and a lot of these bits and pieces end up being something else that we hid but they started working through it – it’s just so cool to see people get together for those kinds of things, there’s something very powerful there and we’ll keep doing it.

Similarly to what happened with the Kosumoth stuff, a couple of them were in zones where it was difficult to be the only person trying to access it, I spent many hours dying to the steam geysers, thanks for that, that was great…


Kosumoth orb six

Is it difficult to react at the right speed to that sort of thing because you don’t want to fix it and then reveal it accidentally to the whole community before they worked that kind of thing out.

A lot of those things we try to be hands off on them because whatever’s in the game if a player figures it out we don’t want to just say ‘oh no that was too easy, that was too hard’, we’ll let it play out and then the next time we do it we have new things that we want to make sure that ‘okay, that was probably too hard, that was probably too challenging’ or ‘that was a piece that felt like it misled players’ so when we do it again let’s try to avoid those things and let’s try to make it better.

It’s hard when people spend so much time doing them to make changes and for them to know the changes exist. Hidden artifact appearances too, if we make a change and we say ‘okay, well we didn’t like how this part of it worked, we didn’t like this particular one’, if we move it somewhere else how does the player know? Do they keep doing that thing that they did? So we try to just let it be, and we try to make it a little more known if we find that there’s a problem.

There was one with the Prot Warrior artifact that you made a subtle post about how you’d changed it and with the Arms one you changed a value, right? And people were very angry about that and you were like ‘look, it was in the game before, but now we’ve made it more likely because it just wasn’t being found.’ Is it just really difficult to find the balance there?

They really are. No matter what we do we’ll get feedback from both sides. We just have to make a gut call of ‘do we think what is happening in the game right now is healthy for the game?’ and if the answer is yes and if a lot of times we realise ‘oh, we made that too common, should we reduce it’ and it’s like that time has passed. The players have found it, it’s cool, it’s awesome that they found it, they’re going to get it, that’s fine. It’s a tough call.

Communication and speed of changes going forward

Legion Q&A

I think you guys have reacted quicker when it comes to Legion than you maybe have in the past and I think Blizzard in general is improving their communication, what’s that been like internally? Have you felt that way as well?

For World of Warcraft specifically we have better tech now to be able to react, so our hotfixing capabilities previously to Legion were actually more limited and our amazing engineers in addition to making our launch super smooth they actually made it so that we can change more than we’ve been able to do. We still have limits on what we can do, but we saw more hotfixes in Legion than I think any other expansion.

We just want to keep making the game better, and that’s one tool we have to make the game better, the other tools we have to make the game better is adding new content, more dungeons, more raids, adding more questlines, adding more stories, adding more artifact appearances, adding more class mounts and we try to balance those two things to make sure that people are having fun now and that they have something to look forward to.

Moving forward, how difficult is it to keep creating stuff that is new. I wrote a couple of times about how Legion feels like World of Warcraft 2 in a lot of ways because you’ve added this stuff on top. How difficult is it going forward to make sure that you can do stuff like adding class mounts and all the rest of it, are you worried that eventually you’re just going to run out of ideas?

Actually not at all, it’s weird. A lot of the features that ended up in Legion were us looking forward and we have a story that we want to tell, we know the direction the story is going but the individual pieces of it like ‘hey, it would be a great time to do this system’ and one of these systems was the scaling world and we were like okay, maybe next expansion we can do this. Actually wait, no we can do it now. Let’s do it now. So we pulled that into Legion and we restructure the world so the different quest zones could scale and you could choose which one you wanted to do, that came out of future stuff.

So we have so many ideas we end up pulling things in now because we don’t know when we’re going to get to them in the future. It’s plans and systems and everything, we just keep going. I think the biggest problem we have is knowing what things to do at any other time, like Legion, if you kind of put a list of all the things we did in Legion, it’s insane. Not all of them are huge like world quests or the scaling world, a lot of them are like little subtle things that we didn’t think we were going to be able to do that actually had a big impacts on the way the game is played. We have pages of those, pages of those that we’d like to do and figuring out when to do them is the real problem.

Well I’m looking forward to it, thanks very much.